Bug 131025 - AX: VoiceOver reads wrong list count when UL is 'clearfixed' and LI are set to inline-block
Summary: AX: VoiceOver reads wrong list count when UL is 'clearfixed' and LI are set t...
Status: NEW
Alias: None
Product: WebKit
Classification: Unclassified
Component: Accessibility (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac (Intel) OS X 10.9
: P2 Normal
Assignee: Nobody
URL:
Keywords: InRadar
Depends on: 131041
Blocks:
  Show dependency treegraph
 
Reported: 2014-04-01 02:54 PDT by Katrin Kerber
Modified: 2014-04-01 09:22 PDT (History)
2 users (show)

See Also:


Attachments
test case (last list in page) (1.00 KB, text/html)
2014-04-01 09:21 PDT, James Craig
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Katrin Kerber 2014-04-01 02:54:15 PDT
When setting list items as inline-block and using the clearfix technique which sets a pseudo element of the UL as display: table, VoiceOver will wrongly read that the list contains 1 item.

QuickCast to show the scenario: http://quick.as/vqpt3da
Comment 1 Radar WebKit Bug Importer 2014-04-01 02:58:47 PDT
<rdar://problem/16483989>
Comment 2 James Craig 2014-04-01 08:54:47 PDT
Hi Katrin, thanks for the screen cast, but we really need a test case. Just attach that list-test.html file you demoed in the video. Thanks.
Comment 3 James Craig 2014-04-01 08:56:03 PDT
Oh, I see the URL attached now. Nm. Thanks.
http://katrinkerber.com/list-test.html
Comment 4 James Craig 2014-04-01 09:20:17 PDT
Katrin's test case because correctly in the WebKit nightly builds. The second list is heuristically determined to be a presentational list (like a layout table) and therefore not exposed as a list at all.

I'm not convinced this bug is gone though. I updated the test case to include an example with explicit roles (that should override the heurisitic presentational determination) and the list is still being exposed as a group, So this bug is currently blocked or masked by bug 131041.
Comment 5 James Craig 2014-04-01 09:21:04 PDT
Created attachment 228291 [details]
test case (last list in page)
Comment 6 James Craig 2014-04-01 09:22:08 PDT
(In reply to comment #4)
> Katrin's test case because correctly…

*behaves* correctly.